Contents
Statistics and data

Distribution: normal, binomial and Poisson from one screen

Skip the data table entirely: name the curve, type the x the question asks about, and read either the height or the area up to it.

8 min read Intermediate updated 2026-11-07

Distribution is the Statistics family with the data taken out. You supply the shape instead: which model, its centre and spread, and the one value the question asks about. The machine then returns either the height of the curve at that x or the area under it up to that x. Nothing is stored between runs, so there is no stale column to clean out first.

Open a distribution and enter the numbers

Keystrokes for a Normal curve with μ = 60 and σ = 5

  1. 1
    MENU

    Open the application menu.

  2. 2
    ▼ ▶

    Move the highlight onto the Distribution icon.

  3. 3
    =

    Open the application.

  4. 4
    ▼ =

    Choose the model. Follow the label rather than a number: the list is Normal, Binomial, Poisson and the rest, and its order differs between EX models.

  5. 5
    ▼ =

    Choose which of the two entries you want - the value at x, or the cumulative area up to x. Every model in this application offers both.

  6. 6
    6 5 =

    The x field: 65, the mark the question asks about.

  7. 7
    5 =

    The spread field: σ = 5.

  8. 8
    6 0 =

    The centre field: μ = 60.

  9. 9
    =

    Confirm and read the result, then press AC to leave. The screen repeats the field names back, so check them once before you write anything down.

fx-991EX Plus 0.
MENU → the Distribution icon → =, then two choices made by label. Every = confirms the current entry and moves on. Rendered from the same layout data the online calculator uses — key positions and silkscreen labels are not stylised.

Normal: the area, not the height

Take marks that follow a Normal curve with μ = 60 and σ = 5. Asking for the area up to 65 is the same question as 'what share of candidates scored 65 or less', and 65 is exactly one σ above the mean.

QuestionScreenHow to check it
P(X ≤ 65)0.841345one σ above the mean: about 84% of the area
P(X ≤ 55)0.158655the mirror image, 1 − the row above
P(55 ≤ X ≤ 65)0.68269the middle strip - the 68% of 68-95-99.7
height at x = 650.048394a density: no width, so no probability
height at x = 600.079788the peak, because the curve is highest at μ
  • P(X > 65) = 1 − 0.841345 = 0.158655. A Normal curve is symmetric about μ, so the right tail above 65 and the left tail below 55 are the same number - a two-second way to catch a σ and μ typed the wrong way round.
  • Subtract before you round: the middle area is the difference of the two full-digit areas the machine is holding, not of the six-decimal numbers printed above.
  • The figures here are rounded to six decimals. Your screen carries more digits, and how many you see is a display setting rather than a calculation.

Binomial and Poisson: counts, not measurements

Both of these work on whole numbers of successes, and both need the height-versus-area choice settled before you type. Ten flips of a fair coin is the smallest case where the whole answer can be checked on paper.

ModelQuestionScreen
Binomial n = 10, p = 0.5exactly 3 heads0.1171875
Binomial n = 10, p = 0.53 heads or fewer0.171875
Poisson λ = 3exactly 2 arrivals0.2240418
Poisson λ = 32 arrivals or fewer0.4231901

The binomial pair is arithmetic: the coefficients of (a + b)¹⁰ run 1, 10, 45, 120, 210 and so on, so P(X = 3) = 120 ÷ 1024 = 0.1171875 and P(X ≤ 3) = (1 + 10 + 45 + 120) ÷ 1024 = 176 ÷ 1024 = 0.171875.

Example Three heads in ten flips, by hand

Input

176 ÷ 1024

Result

0.171875

The denominator is 2¹⁰, one per outcome. If the screen is showing a fraction, S⇔D turns the same result into this decimal.

  • The means come free: a binomial is np = 5 heads, a Poisson is λ = 3 arrivals. An answer a long way from those numbers deserves a second look.
  • The spread is the row after the mean - √(np(1 − p)) = 1.581139 for the coin, √λ = 1.732051 for the arrivals - and it is the same kind of number the Normal curve asked for as σ.
  • Binomial counts successes out of a fixed n; Poisson counts events in a stretch of time or space with no n to point at. Picking the wrong model is the one mistake the machine cannot see.
  • With λ = 3 exactly, P(X = 2) and P(X = 3) are the same 0.2240418: a Poisson curve whose mean is a whole number has two tallest bars, so 'the most likely count' is a question with two answers.

Was this guide helpful?

Tell us what was unclear, missing or wrong. A concrete sentence is more useful than a rating.